A peptide immunotherapy for tumors using HLA class I-binding peptides has been performed for trial purposes. However, judging from the results of peptide immunotherapies obtained to date, administration of a tumor antigen peptide alone is not expected to be so effective (as far as past reports of treatment are concerned, there were almost no cases wherein the ratio of SD in RECIST ratings exceeded 10%) (non-patent references 1, 3).
Hence, to stimulate cytotoxic T lymphocytes (CTLs), a method is widely used wherein a peptide emulsified with Freund's incomplete adjuvant (FIA) is administered (non-patent references 1-3). A peptide suspended in FIA, as is evident from an increase in MHC tetramer-positive CD8 T cells or IFN-γ-secreting cells, allows peptide-specific T cells to proliferate. However, the CD8 T cells that have proliferated are unlikely to be completely activated as an effector; their therapeutic effect has been limited (non-patent references 4-6). In some cases, the peptide suspended in FIA can even cause antigen-specific immunological tolerance (non-patent reference 7). There is another problem of deteriorating the patient's QOL considerably because the FIA long persists under the skin, and also because papules persist for about 2 years and skin induration progresses as the number of administrations increases, although flare dissipates in 1 to 2 months.
In an attempt to increase peptide immunogenicity, many adjuvants have been tested in clinical studies. The purpose of use of these adjuvants is to provide CTLs with an anti-inflammatory environment by activating antigen-presenting cells (APCs) and/or helper T cells (Th). Dendritic cells are used as antigen-presenting cells for peptide pulsation to induce better tumor control (non-patent references 3, 8). Non-methylated deoxy-CpG, Toll-like receptor ligands (non-patent reference 9) and ligands that activate APCs, such as Flt3 (non-patent reference 10), have been introduced. Regarding helper activity, type Th1 responses induce optimal cellular immunity (non-patent references 11, 12). Recombinant cytokines secreted by Th cells and other immunopotentiating cells are used in clinical studies. These cytokines include IL-2 (non-patent reference 13), GM-CSF (non-patent reference 8), IFN-α (non-patent reference 14) and IL-12 (non-patent reference 15). Because commercial products of GMP (Good Manufacturing Practice) grade are available, the cytokines allow easy conduct of test treatment. However, these are no more than capable of replacing some of immune responses, and mainly stimulating either APCs or Th cells only. Additionally, the biological half-lives of cytokines are limited.
For the purpose of activating both APCs and Th cells under more natural conditions, use of the tubercle bacillus cell wall skeleton (BCG-CWS) has been developed (non-patent reference 16). However, BCG-CWS potently activates cellular immunity, so that the onset of open ulcer is unavoidable. For this reason, BCG-CWS cannot be used in patients with decreased immunity, like leukemia patients. BCG-CWS is also difficult to use in repeated immunization. Because peptides are apt to undergo decomposition by proteases in the serum, and disappear within several days, except for those bound to MHC molecules on the cell surface, frequent immunization is necessary. T cells, which essentially respond to a tumor autoantigen protein being a subject of immunological tolerance, are likely to lose activity; the cytotoxic activity thereof weakens unless administered once weekly. Therefore, being difficult to administer repeatedly, BCG-CWS is seriously faulty as an adjuvant for tumor immunotherapy.    non-patent reference 1: Rosenberg, S. A. et al., Nat. Med., vol. 10, p. 909-915 (2004)    non-patent reference 2: Oka, Y. et al., Proc. Natl. Acad. Sci. USA, vol. 101, p. 13885-13890 (2004)    non-patent reference 3: Mosolits, S. et al., Ann. Oncol., vol. 16, p. 847-862 (2005)    non-patent reference 4: Nagorsen, D. et al., Clin. Cancer Res., vol. 12, p. 3064-3069 (2006)    non-patent reference 5: Nencioni, A. et al., Ann. Oncol., vol. 15, p. 153-160 (2004)    non-patent reference 6: Romero, P. et al., Cancer Immunol. Immunother., vol. 53, p. 249-255 (2004)    non-patent reference 7: Toes, R. E. M. et al., Proc. Natl. Acad. Sci. USA, vol. 93, p. 7855-7860 (1996)    non-patent reference 8: Slingluff, C. L. et al., J. Clin. Oncol., vol. 21, p. 4016-4026 (2006)    non-patent reference 9: Speiser, D. E. et al., J. Clin. Invest., vol. 115, p. 739-746 (2005)    non-patent reference 10: Fong, L. et al., Proc. Natl. Acad. Sci. USA, vol. 98, p. 8809-8814 (2001)    non-patent reference 11: Fallarino, F. et al., J Immunol, vol. 165, p. 5495-5501 (2000)    non-patent reference 12: Dredge, K. et al., Cancer Immunol. Immunother., vol. 51, p. 521-531 (2002)    non-patent reference 13: Rosenberg, S. et al., Nat. Med., vol. 4, p. 321-327 (1998)    non-patent reference 14: Di Pucchio, T. et al., Cancer Res., vol. 66, p. 4943-4950 (2006)    non-patent reference 15: Peterson, A. et al., J. Clin. Oncology, vol. 21, p. 2342-2348 (2006)    non-patent reference 16: Nakajima, H. et al., Cancer Immunol. Immunother., vol. 53:, p. 617-624 (2004)